skill-from-research

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.85). The required workflow reads the entire research pack files (outsider-authored free text from paths supplied by the user/pack) and then has the LLM verify/distill claims, so untrusted pack prose can reach the agent’s LLM context via the “read everything, then verify” step.